Studies on the Nature of Chloroplast Lamellae. II. Chemical

II. Chemical Composition and Further Physical Properties of Two Chlorophyll-Protein Complexes* ... Masaaki Takahashi and Elizabeth L. Gross. Biochemis...
1 downloads 9 Views 893KB Size
H

I t 1 ( ' I I 1. hl 1 Y I I< 1

Studies on the Nature of Chloroplast Lamellae. 11. Chemical Composition and Further Physical Properties of Two Chlorophyll-Protein Complexes" J. Philip Thornber.1 John C. Stewart, Mark W. C. Hatton, and J. Leggett Bailey

ABSTRACT: The amino acid and pigment composition of the two major detergent-soluble chlorophyll-proteins (complexes I and 11) of green leaves has been investigated. Complex I (chlorophyll a :b, 12 : 1) contains mainly &carotene and a trace of lutein, whereas complex I1 (chlorophyll a :b, 1.2 :1) contains more xanthophylls than /3-carotene; a considerable proportion of the pigments which are attached to the complexes in riro IS extracted from them during the isolation of the complexes. The protein moieties of the two complexes

T

he lamellar membrane of the chloroplast can be rendered soluble to different extents by a variety of treatments, such as sonication, freeze thawing, or extraction with organic solvents or detergents. It is clear that the dispersion of the lamellae components which accompanies dissolution is not random, but that several significant complexes are released (Anderson and Boardman, 1966; Bril, 1965; Gross et al., 1966; Kahn, 1964; Katoh and San Pietro, 1966; Ogawa et NI., 1966; Sironval et al., 1966; Thornber et ul., 1966, 1967; Vernon et al., 1966; Wessels, 1966a,b). The most important problem in relation to these complexes is whether they correspond to the photochemic~l systems involved in the collection and transfer of energy in photosynthesis. As yet, no complete and unequivocal physical separation of the two systems in an active state has been demonstrated. To this end Anderson and Boardman (1966) have obtained partial separation of the active systems, and Ogawa ct d.(1966) and Sironval et ul. (1966) have completely separated two inactive particles which they believe correspond to the two photochemical systems ; also treatment of chloroplasts with Triton X-100. which inactivates system 11, yields a particle having system I activity (Kahn, 1964; Vernon et id., 1966). In Part I (Thornber e r id., 1967) a fractionation of two photochemically inactive particles was described, and evidence was adduced for the derivation of each of the particles from a n active photochemical system. Apart from the pigment compositions reported . .

~

.... ..~..-

~

~~~

~~

~

~

~~~

- ..

have different compositions, although they both contain a very high proportion of apolar amino acids.

by Ogawa er (11. (1966) and K e et al. (1966), and evidence derived from action spectra (Clayton. 1965), little quantitative information on the individual photochemical reaction centers is available. Thus analysis of our pigmented lipoprotein complexes, even though each may have lost some of the components present in ciuo (Ogawa et al., 1966), can be expected to provide a useful approach t o a description of the photoactive sites in chemical terms. We now wish to report the results of such an analysis. Methods Plant Murerial. Spinach beet (Betti ridgaris, L. ciclcr). spinach (Spinaceci oleraceu), and tobacco (Nicotiuna tubacum) were grown in a greenhouse and the leaves were harvested immediately prior to being homogenized. Oats (Acenci safira var. condor) were grown for 17 days in a greenhouse and the leaves were stored at - 20' overnight before use. Isolation o j Free-Pigment and Pigment-Protein Coniplexes. Full experimental details of the procedure have been given previously (Thornber et a/., 1967). Essentially the method was as follows. Chloroplasts were isolated by differential centrifugation of leaf homogenates, were washed once with water, and then successively extracted five times with 0.5 (w,'v) SDBSl4.05 hi sodium borate (pH 8.0, detergentchlorophyll, 2.5 :1, wlw); this produced five detergent extracts and a small residue. The second and third detergent extracts were resolved into four pigmentcontaining zones by electrophoresis in polyacrylamide

*

2006

From T\\yford Laboratories Ltd., London, N.W. IO. England. Rewired March 7 , 1967. t Present address : Department of Biology, Brookhaven Notional Laboratory, Upton, Long Island, N. Y . 11973.

I H O K N B L K , S ~ L W A r, R H A T

r u ~ A, N D

U A I L L Y

SPO,,, =

9 S for complex I and 2-3 S for complex 11. Both complexes contain carbohydrate material. The only difference between complexes I and I1 prepared from a monocotyledon and those isolated from some dicotyledons was the low 0-carotene content in complex I1 from a monocotyledon. We have contrasted our method for estimation of the pigment composition of the complexes with those used by other workers.

-

__ - - __ _.

-

__

Abbrrv~ntionsused SDBS, sodium dodecylbenzenesulfonate; SDS, sodium dodecyl sulfate, PQ, plastoquinone. 1

VOL.

6,

NO.

7,

J U L Y

1967

gels (Figure 1); complexes I and I1 are pigment-proteins, and complexes I11 and IV are protein-free pigment zones; presumably detergent is attached to each complex. Disks containing each zone were subsequently dissected from the gel columns, and each pigmented complex was eluted from a slurry of the disks through a Sephadex G-200 column. Analytical Ultracentr$igation. Complexes I and I1 were each eluted through a Sephadex G-200 column in order to equilibrate the lipoprotein in a known strength of detergent or in water; equilibration by dialysis took too long. Sedimentation runs were carried out in the An-E rotor of a Spinco Model E ultracentrifuge using cells of 30-mm path length. A rotor speed of 42,040 rpm was used. Samples for analysis contained about 0.5 mg of protein/ml. The viscosities and densities of the solvents were measured and used to correct the determined S values (Schachman, 1957); a partial specific volume of 0.96 was assumed (Chiba, 1960). Amino Acid Analysis. Each of the two chlorophyllproteins was freeze dried and extracted with 80% aqueous acetone to remove pigments and lipids. The delipidized proteins (3-5 mg) were hydrolyzed with constantboiling HC1 at 110" under Nz in tubes sealed under reduced pressure; hydrolysis times ranged from 12 to 72 hr. The amino acid composition of the hydrolysates was obtained using an EEL-Guinness automatic analyzer. Oxidized proteins, prepared by the method of Hirs (1956), were used to determine the content of half-cystine and methionine which were estimated as cysteic acid and methionine sulfone, respectively. After removal of the pigments and lipids from solutions of each complex by treatment with butanol as described by Biggins and Park (1965), the tyrosine: tryptophan molar ratio of the resulting soluble protein was determined by the spectrophotometric method of Beaven and Holiday (1952). Carotenoid Composition. The free-pigment and pigment-protein zones (complexes I-IV, Figure 1) were isolated from 9 ml of detergent extract, containing about 4 mg of chlorophyll and 0.5 mg of carotenoid, within 8 hr of homogenizing the leaves. To the solutions obtained were added three volumes of absolute ethanol and one volume of 10 N NaOH. The mixture was left in the dark at room temperature for 14-16 hr and then diluted with three to four volumes of water. Extraction with peroxide-free ether was carried out until no further pigment was obtained. The combined ether extracts were washed with water until the washings were neutral to litmus and then dried over anhydrous sodium sulfate (Davies, 1965). The whole procedure was carried out in subdued lighting. In order to estimate the total carotenoid, the ether solution was evaporated to dryness and the residue was taken up in a known convenient volume of absolute ethanol. The optical density of this solution at its A,, was measured on a Perkin-Elmer spectrophotometer, Model 137, adjusted so that the optical density at 600 mp was zero. Total carotenoid was calculated using a value for E;& based on the relative amounts of P-carotene and xanthophyll in the solution, this being determined subsequently.

Color

Complex

Green

I - - +

Green

II

ti

Green Yellow

8 1 : A diagrammatic representation of the major complexes separated by electrophoresis on polyacrylamide gel of an SDBS extract of chloroplasts (experimental details are given in Thornber et al., 1967). FIGURE

A sample of the ethanolic solu:ion was examined by thin layer chromatography (tlc) on Kieselgel H. Components of the main yellow bands so obtained were identified by the following criteria: (a) R F values in the two systems, ethyl acetate-carbon tetrachloride (60: 40, v/v) and benzene-methanol (87 :13, v/v); (b) comparison in the same solvents with /3-carotene (Roche Products Ltd., London, W.1) and with the main xanthophylls of grass, isolated by the method of Davies (1965) and confirmed as lutein, violaxanthin, and neoxanthin; (c) color change on the tlc plate when exposed to HCI vapor and after elution from the tlc plate; (d) comparison of the spectra given by each band in petroleum ether (bp 30-60"), chloroform, ethanol, and carbon disulfide with the data of Davies (1965); and (e) spectral shifts on addition of 2 drops of 0.1 N HC1 to the ethanolic solution. The presence or absence of hypsochromic shifts was taken as a guide to the presence or absence of neoxanthin or violaxanthin (Davies, 1965). For the estimation of individual carotenoids a suitable amount of the ethanolic solution was run on 20 X 2.8 cm tlc plates with Kieselgel H as adsorbent and ethyl acetate-carbon tetrachloride (60 :40, v/v) as developing solvent. Immediately after development the plates were scanned in a Chromoscan Model J297 (Joyce Loebl & Co. Ltd., Gateshead, England) using a violet filter (Ilford 622) with approximately level transmittance over the range of 440-456 mp; the quantity of light reflected was recorded. The relative amounts of the main carotenoid bands were calculated by normal integration methods. Estimations were done in duplicate and the average of coincident values was taken. No correction was made for the slightly different extinction coefficients of the carotenoids involved.

2007

NATURE OF CHLOROPLAST LAMELLAE

BIOCHEMISTRY

Chlorophyll Content. The concentrations of chlorophylls a and b were estimated by the method of Mackinney (1941). Nitrogen Content. Nitrogen was estimated by microKjeldahl digestion followed by distillation and titration as described by Chibnall et al. (1943). Carbohydrate Content. Each complex (I and 11) was freeze dried and its pigments and lipids were extracted with 80% aqueous acetone. The residual material was hydrolyzed (2 N H2SO4for 3 hr at 110') and the solution was neutralized with Bio-Deminrolit (carbonate form). The concentrated hydrolysates were chromatographed on Whatman No. 1 paper in ethyl acetate-pyridinewater (8:2:1, vjv), and the monosaccharides were identified by comparison with markers after detection by the method of Trevelyan et al. (1950). IdentiJcation of Other Lipids. Lipids of freeze-dried samples of detergent extracts and complexes I and I1 were extracted with diethyl ether-petroleum ether (bp 100-120") ( l : l , v/v) and examined on thin layer chromatograms by conventional techniques for the presence of quinones, mono- and digalactosyl diglycerides, phospholipid, and sulfolipid. Determination of Manganese. Freeze-dried samples of detergent extracts and complexes I and I1 were digested with H2S04and the charring was cleared by addition of H202and further digestion. The method of determination was based on those of Cornfield and Pollard (1950) and Yuen (1958) and communicated to us by Dr. P. C. de Kock, Macaulay Institute, Aberdeen, Scotland. The procedure differed from the published versions in that a 1 :1 mixture of tetrabase (p,p'-tetramethyldiaminodiphenylmethane) solution and leucomalachite green solution was used to develop the color, which was read at 640 mp, and the quantity of manganese present was calculated from a standard curve. Results

2008

Analysis of the Pigment-Protein Complexes Ultracentrifugation. The SDBS extract, used as the starting solution from which the complexes were to be isolated, was diluted to bring the detergent concentration to 0.1 (w/v) and examined in the ultracentifuge. Two boundaries, one of 10 S and one of 3 S, were observed. Examination of complexes I and I1 in the ultracentrifuge showed that the pigment (and presumably bound detergent) sedimented with a single protein boundary in each case. The s20.wvalues of each complex in various strengths of detergent are presented in Table I and it will be noted that the sedimentation coefficient of complex I is much greater than that of complex 11. If as much of the detergent as possible is removed from the solution by equilibrating the complexes in water, then the lipoproteins aggregate. Complexes I and 11 isolated from the leaves of a monocotyledon have sedimentation coefficients similar to those of spinach beet (Table I). The Protein Moieties. Several samples of the proteins of complexes I and I1 were obtained from different preparations of chloroplasts and were hydrolyzed for

THORNBER,

STEWART,

HATTON,

AND

BAILEY

Sedimentation Coefficients of Complexes I and I1 in Various Strengths of SDBS.

TABLE I:

Complex

sz0.w

I (spinach beet) I (spinach beet) I (spinach beet) I (oat)

0.2 0.02 Water 0.02

8.3 8.8 21.4 8.9

I1 (spinach beet) I1 (spinach beet) I1 (spinach beet) IC (oat)

0.2 0.02 Water 0.02

2.5 3.1 5.7 2.6

a Solutions of SDBS were made up in 0.05 M sodium borate (pH 8.0).

varying lengths of time as indicated in Table 11. The amino acid compositions of the two proteins (Table 11) are very different; the greatest variation between the best analyses occurs in the values for proline, lysine, histidine, glycine, glutamic acid, and aspartic acid. The tyrosine: tryptophan molar ratio was determined as 2.67 for complex I and 2.55 for complex 11. After adjusting all the molar ratios of the amino acids so that the value for half-cystine was one, a minimum molecular weight of each protein was calculated from the best analyses. The figures obtained were 33,300 for complex I and 21,900 for complex 11. The amino acid compositions of the proteins of the two equivalent complexes isolated from chloroplasts of other plant species (spinach, tobacco, and a monocotyledon, oat) are very similar to those of spinach beet chloroplasts (Table 111). The amino acid composition of the material which remains after five successive extractions of the chloroplasts with SDBS solution was also determined but the results were not very consistent. They did, however, indicate that the amino acid composition resembled that of complex I1 more closely than that of complex I (i.e., the values for proline and lysine were high while for histidine they were low). Chlorophyll a : b and Chlorophyll: Protein Ratios of the Pigment-Protein Complexes. The chlorophyll a :b ratio in the two complexes has been reported previously (Thornber et al., 1967); complex I had a mean value of 12:l and complex IIof 1.2: 1. These ratios have nowbeen confirmed by direct estimation of the two chlorophylls after separation by tlc following the procedure of Bacon (1965); traces of pheophytins a and b were observed in each complex. The chlorophyll : protein ratio was determined byestimation of the weight of chlorophyll and of nitrogen in samples of each complex, from which soluble acrylamide contaminants had been removed by gel filtration. The weight of nitrogen was corrected for that contributed by the chlorophyll; no other corrections were made, and all the remaining nitrogen was assumed to be derived from

VOL.

6,

NO.

7,

1967

J U L Y

TABLE 11: Amino Acid

Analysis of Complexes I and 11.~1 Length of Time of Hydrolysis (hr)b ~

~~

Amino Acid

12

24

Half-cystine/ Aspartic acid Methionine Threonine Serine Glutamic acid Proline Glycine Alanine Valine Isoleucine Leucine Tyrosine Phenylalanine Lysine Histidine Arginine Tryptophan

-

-

0.79

0.78

Half-cystinef Aspartic acid Methionine Threonine Serine Glutamic acid Proline Glycine Alanine Valine Isoleucine Leucine Tyrosine Phenylalanine Lysine Histidine Arginine Tryptophan

-

-

48

72

Oxidized. 18 24

A. Complex I 0.03 0.76 0.71 0.84 0.15 0.30 0.12 0.46 0.19 0.04 0.44 0.67 0.63 0.74 0.40 0.40 0.44 0.94 0.91 1.10 0.88 0.88 0.95 0.58 0.49 0.48 0.62 0.57 0.62 1.06 1.06 1.06 0.13 0.25 0.23 0.62 0.62 0.63 0.21 0.27 0.28 0.38 0.45 0.44 0.34 0.31 0.26

-

0.40 0.33 0.82 0.47 0.98 0.87 0.64 0.58 1.06 0.27 0.65 0.30 0.24 0.35

0.48 0.51 0.73 0.44 1.02 0.92 0.56 0.59 1.06 0.22 0.62 0.25 0.33 0.24

-

-

0.94

0.99

0.45 0.33 1.06 0.59 1.15 0.97 0.79 0.59 1.06 0.30 0.60 0.49 0.08 0.44

0.33 0.47 1 .oo 0.86 1.40 1.10 0.79 0.47 1.06 0.02 0.64 0.56 0.12 0.34

B. Complex I1 0.06 0.98 0.99 1.04 0.17 0.24 0.11 0.37 0.12 0.05 0.37 1.00 1.02 0.76 0.79 0.73 1.31 1.31 1.12 1.16 0.69 0.73 0.64 0.49 0.45 0.52 1.06 1.06 1.06 0.28 0.25 0.08 0.66 0.62 0.62 0.52 0.58 0.56 0.13 0.13 0.06 0.33 0.38 0.31

-

-

-

-

-

-

-

-

-

-

-

-

-

Best Anal:

Residuesd

0.03 0.74 0.12 0.37 0.38 0.72 0.52 1.03 0.88 0.53 0.59 1.06 0.01 0.59 0.30 0.26 0.27

0.03 0.77 0.12 0.48 0.51 0.72 0.4.4 1.02 0.90 0.57 0.60 1.06 0.24 0.62 0.27 0.45 0.33 0.09

1 25.7 4 16 17 24 14.7 34 30 19 20 35.3 8 20.7 9 15 11 3

0.05 1.04 0.17 0.34 0.43 1 .oo 0.82 1.53 1.14 0.68 0.51 1.06

0.053 1.01 0.17 0.57 0.47 1 .oo 0.80 1.41 1.15 0.72 0.49 1.06 0.28 0.63 0.57 0.13 0.32 0.11

1 19.1 3.2 10.8 8.9 18.9 15.1 26.4 21.8 13.5 9.3 20.0 5.3 11.9 10.8 2.5 6.0 2.1

-

-

0.59 0.61 0.05 0.27

-

a Expressed as molar ratios after each value (not corrected for losses) had been adjusted so that the value for leucine was constant. * Two samples were used for each set of figures except those of 12 and 18 hr (oxidized) when a single sample was used. c Based on all analyses except the 12-hr values. d Calculated on the basis of one half-cystine residue. e Oxidized proteins were prepared by the method of Hirs (1956). A small peak in the position of cysteic acid was also found with unoxidized samples.

protein. The weight of nitrogen was converted to weight of protein by multiplying by 6.23 (complex I) and 6.48 (complex 11), these figures having been calculated from the amino acid analyses. The mean value of the chlorophyll: protein weight ratio for complex I was 0.1 14 (range 0.100-0.133) and for complex I1 it was 0.075 (range

0.05Cr0.082). From these values and a knowledge of the percentages of complex I and complex I1 in whole lamellae (Thornber et al., 1967) it is possible to calculate an approximate ratio of the weight of chlorophyll in the two complexes; a value of 0.87 (complex I : complex 11) is obtained. Thus if each complex has lost the same per-

NATURE OF

CHLOROPLAST

2009

LAMELLAE

BIOCHLMIST KY

~

~~

TABLE 111: Amino Acid

Composition. of Complexes I and I1 in Different Plant Species. Complex I1

Complex I

Aspartic acid Threonine Serine Glutamic acid Proline Glycine Alanine Valine Isoleucine Leucine Tyrosine Phenylalanine Lysine Histidine Arginine

B. culgarisb

S. oleraceac

Tobaccoc

Oat.

B. culgarid

0.77 0.48 0.51 0.72 0.44 1.02 0.90 0.57 0.60 1.06 0.24 0.62 0.27 0.33 0.33

0.77 0.43 0.38 0.66 0.48 0.99 0.90 0.39 0.57 1.06 0.18 0.60 0.30 0.36 0.29

-

0.42 0.44 0.66 0.49 1.01 0.90 0.56 0.62 1.06 0.18 0.68 0.25 0.34 0.24

0.76 0.45 0.35 0.66 0.36 0.99 0.92 0.48 0.66 1.06 0.22 0.62 0.27 0.36 0.33

1.01 0.33 0.47 1.oo 0.79 1.39 1.13 0.72 0.49 1.06 0.28 0.62 0.57 0.12 0.33

S. oleraceac Tobacco. 0.93 0.25 0.33 0.85 0.76 1.32

0.94 0.30 0.42 0.91 0.78 1.40 1.19 0.66 0.47 1.06 0.23 0.67 0.53 0.09 0.32

1.oo

0.68 0.40 1.06 0.21 0.69 0.60 0.13 0.35

Oatc 0.96 0.37 0.32 0.94 0.69 1.29 1.13 0.70 0.46 1.06 0.27 0.62 0.56 0.10 0.34

Expressed as molar ratio after each value had been adjusted so that the value of leucine was constant. b Best analysis except for histidine, serine, and threonine, which are the values after 24 hr. c Values for single analysis after 24-hr hydrolysis. 0

TABLE IV: Distribution of Carotenoids. in Fractions Obtained from SDBS Extracts of Lamellae of Spinach Beet and Oat.

Proportion of Total Carotenoid Obtained in Each Isolated Complex ( %)

-_ Detergent extract Complex I Complex I1 Free-pigment zone (complex 111) Free-pigment zone (complex IV) Residue a

2010

Spinach Beet

Oat

p-Carotene (+ 0-carotene epoxViola- Neoide) Lutein xanthin xanthin

P-Carotene (+ 6-carotene epoxViola- Neoide) Lutein xanthin xanthin

13 30 52

44.0 92.5 30.8 32.2

37.0 7.0 51.7 49.4

13.1 0 3.2 12.0

5.9 neoxanthin > /3-carotene > violaxanthin and this variation is more serious. There are two main causes of loss of pigments: decomposition, owing largely to the action of light and oxygen, and incomplete recovery from the gel. The latter effect is likely to apply fairly equally, and the order of recovery quoted above probably reflects the stabilities of the compounds concerned. As a first consideration it might be thought that it is for this reason alone that the violaxanthin values are low compared to those quoted by Ke et al. (1966). However, the situation is complicated by two observations. (1) The residue still contains a high proportion of violaxanthin, in spite of being subjected to treatment more drastic than that given to complexes I and 11; and (2) the free-pigment zones (I11 and IV) which are exposed to the same conditions as complexes I and I1 also have a high violaxanthin content. The manner in which we treated our samples immediately prior to estimation of their chlorophyll and carotenoid content differed from that of other investigators (Ke et al., 1966; Ogawa et al., 1966). The chlorophyll content of the complex was estimated directly and within 8 hr of cutting the leaves; the complexes were then treated with alkali which saponifies any esters present and yields the carotenoids free from the other pigments (Davies, 1965). No measurable decomposition of the carotenoids occurred during storage in alkali in the dark. We have favored estimation of carotenoids from densitometer tracings of chromatograms on account of the greater convenience and speed involved, but it should be noted that the results are not exactly the same as those obtained when the elution method is used. The main objection to our method is that light of a range of wavelengths is used and the carotenoids have different A, values; comparison of the two methods suggests that this error is in the range, =t10%. On the other hand, the elution method is slower and in some cases involves an evaporation to dryness, allowing decomposition, which

THORNBER,

STEWART,

HATTON,

AND

BAILEY

certainly does not take place at the same rate for each carotenoid. On balance, we feel that the Chromoscan method is superior. For the reasons given above, we conclude that our methods are preferable and that there is insufficient discrepancy between the methods to account for the variations in carotenoid composition observed between our results and those of other workers; hence the differences are real. The Composition of Chloroplast Lamellae. The nature of the lamellar macromolecular structure is such that the membrane cannot be solubilized by a known technique mild enough to ensure no alteration of individual components. Those techniques which have so far been used have ruptured the membrane to differing extents; with some the resulting particles are large and can only be partially resolved by differential centrifugation, whilst with others much smaller particles are obtained and although these particles may then exist as single entities, they have almost certainly lost many of the compounds with which they were associated in uiuo. We prefer the second of these two states, since the products are more amenable to fractionation by classical protein techniques and the purity of fractions can be examined directly. Thornber et al. (1967) showed that two chlorophyllprotein complexes could be purified from chloroplasts dissociated by the anionic detergents, SDBS and SDS; it was also shown that these two components contained over 75% of the protein of whole lamellae. After isolation, complex I is larger than complex 11. This is demonstrated by their sedimentation coefficients (Table I) and their movement in polyacrylamide gel during electrophoresis (Figure 1) ; the difference in electrophoretic mobility is almost certainly due to a difference in size rather than in net charge, since no electrophoretic separation of the two complexes occurs on a solid support that does not have the fine molecular-sieving properties of polyacrylamide gel. Previous reports of the amino acid composition of chloroplast lamellae have been limited to studies by Bailey et al. (1966), Criddle (1966), and Weber (1962). From an acetone powder of lamellae Criddle (1966) has isolated a structural protein which represents over 40% of the total protein of the chloroplast. Since this protein constitutes such a great proportion of the lamellae protein, it is likely to be related to one or possibly both of our proteins; however, there is little correspondence between the amino acid composition of the structural protein and that of either of our proteins. However this does not completely eliminate a relation, since Criddle (1966) reported only one analysis of his protein and that was after 12-hr hydrolysis; it might, therefore, be possible to relate the structural protein with our components after further investigation of its amino acid composition; complex I has a minimum molecular weight similar to that of the structural protein. The combined analyses of the proteins of our two complexes show a mean molar ratio for each amino acid residue corresponding very closely to the values reported for whole lamellae protein (Bailey et al., 1966; Weber, 1962). This very probably confirms our earlier observation that the proteins of complexes I and I1 make up the bulk of the lamellae pro-

VOL.

6,

NO.

7, J U L Y 1 9 6 7

tein. It is clear (Table 11) that the two proteins are of a different composition and minimum molecular weight, Hatch (1965) has shown that the physical characteristics of a protein can be correlated with its amino acid composition; the properties of the proteins of complexes I and I1 are in accord with this hypothesis. If the polar: apolar ratio of the amino acid residues in each protein is calculated as described by Hatch (1965), the resulting values (complex I, 0.91; complex 11, 1.02) indicate an unusually high proportion of apolar amino acids. The behavior of the two proteins is typical of proteins containing a high proportion of apolar amino acids, for example, they aggregate when their lipid is removed, and they interact with other apolar material giving rise to lipoprotein membranes. Van Wyk (1966) and Weber (1963) have shown that chloroplast lamellae contain a structural glycoprotein of the lamellae protein. We which represents 25-35 detect monosaccharides in hydrolysates of both of our delipidized proteins, but in addition to the sugars observed by Van Wyk (1966), our proteins contain arabinose; this is mainly in complex I. The function of the carotenoid pigments in photosynthesis continues to be a matter for speculation. They have been considered suitable for protection of the chlorophylls against photooxidation (Stanier and Cohen-Bazire, 1957) and as photoreceptors in an accessory role (Whittingham, 1965). The xanthophylls specifically have been implicated in the Hill reaction (Sapozhnikov et al., 1964) but no satisfactory explanation of this involvement has been provided, and in fact Yamamoto et al. (1962) were unable to demonstrate any direct connection. The present work can make only a limited contribution to any hypothesis of carotenoid function, since the particles examined were inactive, but two results are of interest. First, complex I contains very little xanthophyll (Table IV), whereas complex 11, derived from the system responsible for the Hill reaction, contains considerably more xanthophyll than p-carotene. Second, a consideration of the results of Ke et a/. (1966) suggests that high values for the chlorophyll a:b ratio for photochemical system I are accompanied by high values for the P-carotene:xanthophyll ratio, and our results lend support to this generalization. Electrophoresis on polyacrylamide gel of any detergent extract yields, in addition to the pigment-protein zones, two free-pigment zones (Figure 1). It can be seen from Table IV that these zones contain a large amount of the carotenoid of the detergent extract; furthermore, their carotenoid composition is as definite as that of any other fraction, and the carotenoids d o not appear to have been released from one particular pigment-protein. These free-pigment zones may be present because the pigments they contain are not bound to any protein in uico or because their binding to protein is weak or their position relatively “external.” Continued exposure of the isolated complexes I and I1 to detergent, which extracts more pigment, may eventually produce chlorophyll-proteins from which no further pigments can be extracted by detergent. Lastly there is the problem of the nature of the re-

sidual material after detergent extractions of chloroplasts; the least readily extracted of the complexes is complex I1 (Thornber et al., 1967) and, therefore, it would be expected that unless another detergentinsoluble pigment-protein is present, the amino acid and pigment composition of the residue (containing some 1.3 of the lamellar protein and 4 of the chlorophyll) would reflect that of complex 11; the results do not substantiate this and thus the residual material must contain other minor components, the nature and function of which requires further investigation. The major conclusions about the two pigment-proteins studied in this work can be summarized as follows. (a) The major portion (over 7573 of the lamellae protein is present in the two complexes. (b) The complexes are different with respect to their protein composition and pigment content. (c) Complexes I and I1 are very probably derived from photochemical systems I and 11, respectively. (d) Complex I is more readily extracted from the lamellar matrix. (e) Complex I is larger than complex 11. (f) Some protein-poor residual material remains after detergent extractions. Hence these observations support a concept of two distinct photochemical systems containing different protein and pigment assemblies. How the two complexes might be accommodated into the models of the lamellae proposed by many investigators is very largely a matter of conjecture. It is tempting to postulate that each complex is a dissociated form of one of the two types of particle observed on different sides of the lamellar membrane (Branton and Park, 1965; Park, 1965, 1966; Muhlethaler et al., 1965); the residue may be some extracted derivative of the supporting matrix. An investigation by electron microscopy of lamellae treated with detergent should indicate whether this postulation is plausible. Acknowledgments We thank Miss Carol A. Smith for her skillful technical assistance. We also wish to thank Dr. J. F. Pennock, Biochemistry Department, University of Liverpool, England, for samples of ubiquinone 10, plastoquinone 9, a-tocopherolquinone, and vitamin kl; Roche Products Ltd., London, England, for samples of plastoquinone 9 and ubiquinone 10; and Professor A. A. Benson, Scripps Institute of Oceanography, La Jolla, Calif., for a sample of sulfolipid. We are greatly indebted t o Professor T. W. Goodwin and his colleagues a t the University College of Wales, Aberstwyth, Wales, for valuable discussions about estimation of carotenoids. References Anderson, J. M., and Boardman, N. K. (1 966), Biochim. Biophys. Acta 112,403. Bacon, M. F. (1965), J. Chromatog. 17,322. Bailey, J. L., Thornber, J. P., and Whyborn, A, G. (1966), in Biochemistry of Chloroplasts, Vol. 1, Goodwin, T. W., Ed., London, Academic, p 243. Beaven, G. H., and Holiday, E. R. (1952), Aduan.

2013

NATURE OF CHLOROPLAST LAMELLAE

BIOCHEMISTRY

Protein Chem. 7,319. Biggins, J., and Park, R. B. (1965), Plant Physiol. 40, 1109. Branton, D., and Park, R. B. (1965), Plant Physiol., xxviii. Bril, C. (1965), Carnegie Znst. Washington Yearbook 64, 370. Chiba, Y. (1960), Arch. Biochem. Biophys. 90,294. Chibnall, A. C., Rees, M. W., and Williams, E. F. (1943), Biochem. J. 37,354. Clayton, R. K . (1965), Molecular Physics in Photosynthesis, New York, N. Y., Blaisdell. Cornfield, A. H., and Pollard, A. G. (1950), J . Sci. FoodAgr. I, 107. Criddle, R. S. (1966), in Biochemistry of Chloroplasts, Vol. 1, Goodwin, T. W., Ed., London, Academic, p 203. Davies, B. H. (1965), in Chemistry and Biochemistry of Plant Pigments, Goodwin, T. W., Ed., London, Academic, p 489. Gross, J., Shefner, A., and Becker, M. (1966), Nature 209,615. Hatch, F. (1 965), Nature 206,777. Hirs, C. H. W. (1956), J . Biol. Chem. 219,611. Kahn, J. S. (1964), Biochim. Biophys. Acta 79,234. Katoh, S., and San Pietro, A. (1966), J. Biol. Chem. 241,3575. Ke, B., Seliskar, C., and Breeze, R. (1966), Plant Physiol. 41,1081. Mackinney, G. (1941), J. B i d . Chem. 140,315. Muhlethaler, K., Moor, H., and Szarkowski, J. W. (1965), Planta 67,305. Ogawa, T., Obata, F., and Shibata, K . (1966), Biochim. Biophys. Acta 112,223. Park, R. B. (1965), J. CellBiol. 27,151. Park, R. B. (1966), in The Chlorophylls, Vernon, L. P.,

2014

THORNBER,

STEWART,

HATTON,

AND

BAILEY

and Seely, G. R., Ed., London, Academic, p 283. Sapozhnikov, D. I., Alkhazov, D. G., Eidelman, Z. M., Bazhanova, N. V., Lemberg, I. K., Maslova, T. G., Girshin, A. B., Popova, I. A., Saakov, V. S., Popova, 0. F., and Shiryaeva, G. A. (1964), Dokl. Akad. Nauk. SSSR Botan. Sci. Sect. 154,20. Schachman, H. K . (1957), Methods Enzymol. 4,32. Sironval, C., Clijsters, H., Michel, J.-M., Bronchart, R., and Michel-Wolwertz, R. M. (1966), Proc. 2nd Western Europe Conf. Photosyn. Zeist, Netherlands, Rotterdam, Donker, Ad., p 111. Stanier, R. Y., and Cohen-Bazire, G. (1957), 7th Symp. SOC. Gen. Microbiol., London, Cambridge University, p 56. Thornber, J. P., Gregory, R. P. F., Smith, C. A,, and Bailey, J. L. (1967), Biochemistry (in press). Thornber, J. P., Smith, C. A., and Bailey, J. L. (1966), Biochem. J . 100,14P. Trevelyan, W. E., Proctor, D. P., and Harrison, J. S. (1950), Nature 166,444. Van Wyk, D. (1966),Z. Naturforsch. 21b, 700. Vernon, L. P., Shaw, E., and Ke, B. (1966), J. B i d . Chem. 241,4101. Weber, P. (1962), 2.Naturforsch. 176,683. Weber, P. (1963),2. Naturforsch. 18b, 1105. Wessels, J. S. C. (1966a), Proc. 2nd Western Europe Conf. Photosyn. Zeist, Netherlands, Rotterdam, Donker, Ad., p 129. Wessels, J. S. C. (1966b), Biochim. Biophys. Acta 126, 581. Whittingham, C. P. (1965), in Chemistry and Biochemistry of Plant Pigments, Goodwin, T. W., Ed., London, Academic, p 360. Yamamoto, H. Y., Chichester, C. O., and Nakayama, T. 0. M. (1962), Photochem. Photobiol. I , 53. Yuen, S. H. (1958), Analyst 83, 350.